CN109447154A - Picture similarity detection method, device, medium and electronic equipment - Google Patents
Picture similarity detection method, device, medium and electronic equipment Download PDFInfo
- Publication number
- CN109447154A CN109447154A CN201811270419.3A CN201811270419A CN109447154A CN 109447154 A CN109447154 A CN 109447154A CN 201811270419 A CN201811270419 A CN 201811270419A CN 109447154 A CN109447154 A CN 109447154A
- Authority
- CN
- China
- Prior art keywords
- picture
- similarity
- detected
- preset
- algorithm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
Abstract
The embodiment of the invention provides a kind of picture similarity detection method, device, medium and electronic equipments, this method comprises: obtaining picture to be detected, the picture to be detected is compared with the reference base picture in preset reference map valut using the first alignment algorithm, obtains the first similarity;When first similarity meets preset similarity threshold condition, complete to detect the similarity of the picture to be detected;When first similarity is unsatisfactory for the preset similarity threshold condition, the picture to be detected and the reference base picture in the preset reference map valut are compared again using the second alignment algorithm, the second similarity is obtained, and according to second similarity determines whether that the picture to be detected is compared with the reference base picture in the preset reference map valut again;Wherein, the comparison accuracy of the second alignment algorithm is higher than the precision of first alignment algorithm.The technical solution of the embodiment of the present invention can be improved the detection accuracy of picture similarity.
Description
Technical field
The present invention relates to image processing technology, in particular to a kind of picture similarity detection method, device,
Medium and electronic equipment.
Background technique
With the development of internet technology with the promotion of development of games technology, the display for exporting picture in game is required
Also it is gradually increased.In general, needing to first pass through resource before the picture output in game is shown and checking system, and use image
The figure of output is compared recognizer with the figure in correct picture library, when the similarity testing result meets the requirements, ability
Enough output uses.And the resource checks that the core of system is exactly image similarity detection algorithm.
In the prior art, the similarity of picture is detected, it will generally by the single color histogram nomography of use
The comparison of picture piecemeal, obtains similarity testing result.Therefore, not only the speed of service is slower for this detection method, and efficiency is lower, and
And accuracy rate is lower, can not effectively detect to a large amount of pictures.
Therefore, picture similarity detection method efficiency and accuracy rate in the prior art are to be improved.
It should be noted that information is only used for reinforcing the reason to background of the invention disclosed in above-mentioned background technology part
Solution, therefore may include the information not constituted to the prior art known to persons of ordinary skill in the art.
Summary of the invention
The embodiment of the present invention is designed to provide a kind of picture similarity detection method, and then at least to a certain extent
Overcome the lower defect of picture similarity detection method efficiency and accuracy rate in the prior art.
Other characteristics and advantages of the invention will be apparent from by the following detailed description, or partially by the present invention
Practice and acquistion.
According to a first aspect of the embodiments of the present invention, a kind of picture similarity detection method is provided, comprising: obtain to be checked
Mapping piece is compared the picture to be detected and the reference base picture in preset reference map valut using the first alignment algorithm
It is right, obtain the first similarity;When first similarity meets preset similarity threshold condition, complete to described to be detected
The similarity of picture detects;When first similarity is unsatisfactory for the preset similarity threshold condition, the second ratio is utilized
The picture to be detected and the reference base picture in the preset reference map valut are compared by algorithm again, it is similar to obtain second
Degree, and being determined whether according to second similarity again will be in the picture to be detected and the preset reference map valut
Reference base picture is compared;Wherein, the comparison accuracy of the second alignment algorithm is higher than the precision of first alignment algorithm.
In some embodiments of the invention, aforementioned schemes are based on, determine whether again to according to second similarity
The picture to be detected is compared with the reference base picture in the preset reference map valut, comprising: judges second phase
Whether meet the preset similarity threshold condition like degree;When second similarity meets the preset similarity threshold
When condition, complete to detect the similarity of the picture to be detected;When second similarity is unsatisfactory for preset condition, utilize
Third alignment algorithm compares the picture to be detected and the reference base picture in the preset reference map valut again, obtains the
Three similarities;The similarity testing result to the picture to be detected is determined according to the third similarity;Wherein, third compares
The comparison accuracy of algorithm is higher than the precision of second alignment algorithm.
In some embodiments of the invention, aforementioned schemes are based on, first alignment algorithm is the picture of the first precision
Hash algorithm, second alignment algorithm are the picture hash algorithm of the second precision, and the third alignment algorithm is peak value noise
Than the combination of algorithm and structural similarity algorithm.
In some embodiments of the invention, aforementioned schemes are based on, using the first alignment algorithm by the picture to be detected
Be compared with the reference base picture in preset reference map valut, obtain the first similarity, comprising: will the picture to be detected and
Reference base picture in the preset reference map valut narrows down to same order, and is converted to corresponding gray scale picture;To institute
It states the progress discrete cosine transform of gray scale picture and obtains coefficient matrix;The coefficient matrix is handled, picture to be detected is generated
Fingerprint and reference base picture fingerprint;The picture fingerprint to be detected and the reference base picture fingerprint are compared, the first similarity is obtained.
In some embodiments of the invention, aforementioned schemes are based on, using the second alignment algorithm by the picture to be detected
It is compared again with the reference base picture in the preset reference map valut, obtains the second similarity, comprising: the picture to be detected
It is split with the reference base picture in the preset reference map valut, generates corresponding at least one segmentation picture;To described
At least one segmentation picture runs first alignment algorithm again, obtains the second similarity.
In some embodiments of the invention, aforementioned schemes are based on, using third alignment algorithm by the picture to be detected
It is compared again with the reference base picture in the preset reference map valut, obtains third similarity, comprising: to the mapping to be checked
Reference base picture in piece and the preset reference picture library runs Y-PSNR algorithm, obtains Y-PSNR calculated result;
To the reference base picture operating structure Similarity Algorithm in the picture to be detected and the preset reference picture library, structure phase is obtained
Like property calculated result;By the Y-PSNR calculated result with the structural similarity calculated result respectively multiplied by respective pre-
If weight, third similarity is obtained.
In some embodiments of the invention, aforementioned schemes are based on, picture to be detected is obtained, comprising: are based on edge detection
Algorithm carries out edge detection to original image, obtains the outer layer coordinate of original image;Based on the outer layer coordinate pair original image
It is cut, obtains picture to be detected.
According to a second aspect of the embodiments of the present invention, a kind of picture similarity detection apparatus is provided, comprising: first compares
Module will be in the picture to be detected and preset reference map valut using the first alignment algorithm for obtaining picture to be detected
Reference base picture be compared, obtain the first similarity, when first similarity meets preset similarity threshold condition,
It completes to detect the similarity of the picture to be detected;Second comparison module, described in being unsatisfactory for when first similarity
When preset similarity threshold condition, using the second alignment algorithm by the picture to be detected and the preset reference map valut
In reference base picture compare again, obtain the second similarity, and according to second similarity determine whether again will it is described to
Detection picture is compared with the reference base picture in the preset reference map valut, wherein the comparison essence of the second alignment algorithm
Degree is higher than the precision of first alignment algorithm.
According to a third aspect of the embodiments of the present invention, a kind of computer-readable medium is provided, computer is stored thereon with
Program realizes the picture similarity detection side as described in first aspect in above-described embodiment when described program is executed by processor
Method.
According to a fourth aspect of the embodiments of the present invention, a kind of electronic equipment is provided, comprising: one or more processors;
Storage device, for storing one or more programs, when one or more of programs are held by one or more of processors
When row, so that one or more of processors realize the picture similarity detection side as described in first aspect in above-described embodiment
Method.
Technical solution provided in an embodiment of the present invention can include the following benefits:
In the technical solution provided by some embodiments of the present invention, on the one hand, by obtaining picture to be detected, utilize
The picture to be detected is compared first alignment algorithm with the reference base picture in preset reference map valut, obtains the first phase
Like degree, when first similarity meets preset similarity threshold condition, the similarity to the picture to be detected is completed
Detection can complete the similarity detection to picture to be detected within a short period of time, improve the efficiency of picture detection;Another party
Face is higher than the first alignment algorithm using precision when first similarity is unsatisfactory for the preset similarity threshold condition
The second alignment algorithm the picture to be detected and the reference base picture in the preset reference map valut are compared again, obtain
Second similarity, and determined whether again according to second similarity by the picture to be detected and the preset reference map
Reference base picture in valut is compared, can reduce picture detection rate of false alarm, improve picture similarity detection precision with
And accuracy rate.
It should be understood that above general description and following detailed description be only it is exemplary and explanatory, not
It can the limitation present invention.
Detailed description of the invention
The drawings herein are incorporated into the specification and forms part of this specification, and shows and meets implementation of the invention
Example, and be used to explain the principle of the present invention together with specification.It should be evident that the accompanying drawings in the following description is only the present invention
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.In the accompanying drawings:
Fig. 1 diagrammatically illustrates the flow chart of the picture similarity detection method of embodiment according to the present invention;
Fig. 2 diagrammatically illustrates the flow chart of picture similarity detection method according to another embodiment of the present invention;
Fig. 3 diagrammatically illustrates the flow chart of the picture similarity detection method of another embodiment according to the present invention;
Fig. 4 diagrammatically illustrates the flow chart of picture similarity detection method according to still another embodiment of the invention;
Fig. 5 diagrammatically illustrates the picture similarity detection apparatus block diagram of embodiment according to the present invention;
Fig. 6 shows the structural schematic diagram for being suitable for the computer system for the electronic equipment for being used to realize the embodiment of the present invention.
Specific embodiment
Example embodiment is described more fully with reference to the drawings.However, example embodiment can be with a variety of shapes
Formula is implemented, and is not understood as limited to example set forth herein;On the contrary, thesing embodiments are provided so that the present invention will more
Fully and completely, and by the design of example embodiment comprehensively it is communicated to those skilled in the art.
In addition, described feature, structure or characteristic can be incorporated in one or more implementations in any suitable manner
In example.In the following description, many details are provided to provide and fully understand to the embodiment of the present invention.However,
It will be appreciated by persons skilled in the art that technical solution of the present invention can be practiced without one or more in specific detail,
Or it can be using other methods, constituent element, device, step etc..In other cases, it is not shown in detail or describes known side
Method, device, realization or operation are to avoid fuzzy each aspect of the present invention.
Block diagram shown in the drawings is only functional entity, not necessarily must be corresponding with physically separate entity.
I.e., it is possible to realize these functional entitys using software form, or realized in one or more hardware modules or integrated circuit
These functional entitys, or these functional entitys are realized in heterogeneous networks and/or processor device and/or microcontroller device.
Flow chart shown in the drawings is merely illustrative, it is not necessary to including all content and operation/step,
It is not required to execute by described sequence.For example, some operation/steps can also decompose, and some operation/steps can close
And or part merge, therefore the sequence actually executed is possible to change according to the actual situation.
In the prior art, the similarity of picture is detected, it will generally by the single color histogram nomography of use
The comparison of picture piecemeal, obtains similarity testing result.Therefore, not only the speed of service is slower for this detection method, and efficiency is lower, and
And accuracy rate is lower, can not effectively be detected to a large amount of pictures.
In the present embodiment, a kind of picture similarity detection method is provided firstly, is overcome at least to a certain extent existing
There are the picture similarity detection method efficiency and the lower defect of accuracy rate in technology.
Fig. 1 diagrammatically illustrates a kind of flow chart of picture similarity detection method according to an embodiment of the present invention, the inspection
The executing subject of survey method can be the server detected to picture similarity.
With reference to Fig. 1, picture similarity detection method according to an embodiment of the invention the following steps are included:
Step S101 obtains picture to be detected, using the first alignment algorithm by the picture to be detected and preset benchmark
Reference base picture in picture library is compared, and obtains the first similarity;
Step S102 is completed when first similarity meets preset similarity threshold condition to described to be detected
The similarity of picture detects;
Step S103 utilizes the second ratio when first similarity is unsatisfactory for the preset similarity threshold condition
The picture to be detected and the reference base picture in the preset reference map valut are compared by algorithm again, it is similar to obtain second
Degree, and being determined whether according to second similarity again will be in the picture to be detected and the preset reference map valut
Reference base picture is compared;
Wherein, the comparison accuracy of the second alignment algorithm is higher than the precision of first alignment algorithm.
In technical solution provided by embodiment shown in Fig. 1, on the one hand, by obtaining picture to be detected, utilize first
The picture to be detected is compared alignment algorithm with the reference base picture in preset reference map valut, and it is similar to obtain first
Degree is completed to examine the similarity of the picture to be detected when first similarity meets preset similarity threshold condition
It surveys, the similarity detection to picture to be detected can be completed within a short period of time, improve the efficiency of picture detection;Another party
Face is higher than the first alignment algorithm using precision when first similarity is unsatisfactory for the preset similarity threshold condition
The second alignment algorithm the picture to be detected and the reference base picture in the preset reference map valut are compared again, obtain
Second similarity, and determined whether again according to second similarity by the picture to be detected and the preset reference map
Reference base picture in valut is compared, can reduce picture detection rate of false alarm, improve picture similarity detection precision with
And accuracy rate.
The specific implementation process of each step in Fig. 1 is described in detail below:
In step s101, obtain picture to be detected, using the first alignment algorithm by the picture to be detected with it is preset
Reference base picture in reference map valut is compared, and obtains the first similarity.
In the exemplary embodiment, it before obtaining picture to be detected, needs first to pre-process original image, in advance
Processing is the processing for carrying out being carried out before feature identification, segmentation and matching to original image.Pretreated main purpose is to eliminate
Unrelated information in image, restores useful real information, enhances detectability for information about and simplifies number to the maximum extent
According to improve the reliability of feature identification, image segmentation, matching and identification.
Illustratively, the pretreatment stage in the present invention mainly includes that picture cuts processing and picture scaling processing.Picture
The process for cutting processing may is that the outermost layer coordinate that object in original image is found using edge detection algorithm.Namely
Say, the coordinate of a pixel if (X, Y), then find out it is all indicate roles pixels in X, the maximum and minimum value of Y,
With (Xmin, Ymin), (Xmax, Ymin), (Xmin, Ymax), (Xmax, Ymax) four pixels are the vertex of rectangle, to original
Image is cut, the image after being cut, i.e. completion picture cutting processing;The purpose of picture scaling processing is by original graph
Reference base picture size adjustment in piece and reference map valut is consistent, so as to subsequent detection and comparison.By to the pre- of original image
Processing, can get picture to be detected.
In the exemplary embodiment, referring to shown in Fig. 2, Fig. 2 diagrammatically illustrates picture according to an embodiment of the present invention
Similarity detection algorithm flow chart, specifically illustrates and obtains the flow chart of the first similarity using the first alignment algorithm, below in conjunction with
Fig. 2 explains step S101.
In step s 201, the picture to be detected and the reference base picture in the preset reference map valut are narrowed down to
Same order, and be converted to corresponding gray scale picture.
In the exemplary embodiment, the first alignment algorithm is the picture hash algorithm of the first precision, i.e. the first precision
Perceptual hash algorithm, its effect are to generate " fingerprint " character string to every image, then compare the fingerprint of different images,
As a result closer, just illustrate that image is more similar.It is possible, firstly, to first by the benchmark in picture to be detected and preset reference map valut
Picture narrows down to 8*8 size, in total 64 pixels, in turn, converts 64 grades of gray scale pictures for the picture after diminution.
In step S202, discrete cosine transform is carried out to the gray scale picture and obtains coefficient matrix.
In the exemplary embodiment, above-mentioned 64 grades of gray scale pictures can be subjected to discrete cosine transform (Discrete
Cosine Transform, referred to as: DCT), obtain the coefficient matrix of 32*32.The discrete cosine transform of image is widely used in image
Compression.Discrete cosine transform is carried out to original image, DCT coefficient energy is concentrated mainly on the upper left corner after transformation, remaining big portion
Divide coefficient that there is the characteristic suitable for compression of images close to zero, DCT.The direct transform formula of two-dimension discrete cosine transform isIn the compression of image, N mono-
As take 8;Work as u, when v=0;Coefficient C (u),In the case of other, C (u),According to discrete cosine
Transformation for mula, the coefficient matrix of available 32*32.
In step S203, the coefficient matrix is handled, generates picture fingerprint to be detected and reference base picture fingerprint.
In the exemplary embodiment, the coefficient matrix of above-mentioned 32*32 is done into diminution processing, obtains the discrete cosine of 8*8
Matrix, and calculate the average value of the 8*8 discrete cosine matrix.In turn, 64 cryptographic Hash of setting 0 or 1 will be greater than average
The numerical value of value is set as " 1 ", and the numerical value for being less than average value is set as " 0 ", then, together by all combinations of values, that is, is constituted
One 64 character string, 64 character strings are the picture fingerprint.
In step S204, the picture fingerprint to be detected and the reference base picture fingerprint are compared, the first similarity is obtained.
In the exemplary embodiment, by comparing the fingerprint of picture and reference base picture to be detected, i.e. two pictures operation
64 character strings generated after the picture hash algorithm of first precision, can be obtained the first of picture to be detected and reference base picture
Similarity.
Continue to refer to figure 1, obtain the first similarity after, in step s 102, when first similarity meet it is preset
When similarity threshold condition, complete to detect the similarity of the picture to be detected.
In the exemplary embodiment, after obtaining the first similarity, if the first similarity meets preset similarity
Threshold condition, such as: preset similarity threshold condition are as follows: similar threshold value a, dissimilar threshold value is b, when the first similarity is small
Then detect that picture to be detected is similar to the reference base picture in preset reference picture library when being equal to a, at this point it is possible to will be to mapping
Piece is moved to similar pictures file, completes to detect the similarity of picture to be detected;When the first similarity is more than or equal to b,
Then detect that the reference base picture in picture to be detected and preset reference picture library is dissimilar, at this point it is possible to picture tune to be detected
With difference algorithm, difference nomography is mainly the difference marked between picture to be detected and reference base picture, bigger to diversity factor
Place infused with red collimation mark, there is the place of different to be highlighted, do not have discrepant place to carry out translucent processing.Finally will
It is moved to wrong catalogue to mapping, and generates disparity map simultaneously, is put into difference drawings list, is completed to the similar of picture to be detected
Degree detection.
In step s 103, when first similarity is unsatisfactory for the preset similarity threshold condition, is utilized
Two alignment algorithms compare the picture to be detected and the reference base picture in the preset reference map valut again, obtain second
Similarity, and determined whether again according to second similarity by the picture to be detected and the preset reference map valut
In reference base picture be compared.
It should be noted that the comparison accuracy of the second alignment algorithm is higher than the precision of first alignment algorithm.
In the exemplary embodiment, referring to step S102 relevant explanation, when the first similarity be greater than a and be less than b, i.e.,
First similarity is between similar threshold value and dissimilar threshold value, then can use the second alignment algorithm and carry out to picture to be detected
Further detection.Second alignment algorithm is the picture hash algorithm of the second precision.The picture hash algorithm of second precision and
The difference of the picture hash algorithm of one precision is: before the perceptual hash algorithm for running the second precision, needing to figure to be detected
Piece is split with the reference base picture in preset reference map valut, generates corresponding at least one segmentation picture;To it is described extremely
Few segmentation picture runs the perceptual hash algorithm of the first precision again, obtains the second similarity.
In the exemplary embodiment, referring to shown in Fig. 3, Fig. 3 diagrammatically illustrates picture according to an embodiment of the present invention
Similarity detection algorithm flow chart, specifically illustrate according to the second similarity determine whether again by the picture to be detected with it is described
The flow chart that reference base picture in preset reference map valut is compared explains step S103 below in conjunction with Fig. 3.
In step S301, judge whether second similarity meets the preset similarity threshold condition.
In the exemplary embodiment, firstly, preset similar threshold value can be c, dissimilar threshold value is d.
In step s 302, it when second similarity meets the preset similarity threshold condition, completes to institute
State the similarity detection of picture to be detected.
In the exemplary embodiment, picture to be detected and default base are then detected when the second similarity is less than or equal to c
Reference base picture in quasi- picture library is similar, at this point it is possible to which picture to be measured is moved to similar pictures file, completes to be detected
The similarity of picture detects;When the first similarity is more than or equal to d, then detect in picture to be detected and preset reference picture library
Reference base picture it is dissimilar, at this point it is possible to call difference algorithm to picture to be detected, difference nomography mainly marks to be detected
Difference between picture and reference base picture, the place bigger to diversity factor are infused with red collimation mark, have the place of different highlighted
Display does not have discrepant place to carry out translucent processing.It will finally be moved to wrong catalogue to mapping, and generate difference simultaneously
Figure, is put into difference drawings list, completes to detect the similarity of picture to be detected.
It, will be described using third alignment algorithm when second similarity is unsatisfactory for preset condition in step S303
Picture to be detected compares again with the reference base picture in the preset reference map valut, obtains third similarity.
In the exemplary embodiment, when the first similarity is greater than c and is less than d, i.e. the second similarity is in similar threshold value
Between dissimilar threshold value, then it can use third alignment algorithm and picture to be detected further detected.Third compares
Algorithm is the combination of Y-PSNR algorithm and structural similarity algorithm.
In the exemplary embodiment, referring to shown in Fig. 4, Fig. 4 diagrammatically illustrates picture according to an embodiment of the present invention
Similarity detection algorithm flow chart, specifically illustrates and obtains the flow chart of third similarity using third alignment algorithm, below in conjunction with
Fig. 4 explains step S303.
In step S401, peak value is run to the reference base picture in the picture to be detected and the preset reference picture library
Signal-to-noise ratio (SNR) Algorithm obtains Y-PSNR calculated result.
In the exemplary embodiment, using Y-PSNR (Peak Signal to Noise Ratio, referred to as:
PSNR) index is objectively evaluated as image.The PSNR of one signal is its maximum power and the expression precision that may influence it
The ratio of noise power, specific calculation formula are as follows:
Wherein, MSE indicates mean square error (Mean Square Error, each data mistake of present image X and reference picture Y
The average of difference square), H, W are respectively the height and width of image;N is the bit number of every pixel, generally takes 8, i.e. pixel ash
Spend order be 256, PSNR unit be dB, PSNR value is bigger, with regard to representative image be distorted it is fewer, i.e., picture to be detected with it is preset
The similarity degree of reference base picture in reference map valut is higher.By to the benchmark in picture to be detected and preset reference picture library
After picture runs Y-PSNR algorithm, available Y-PSNR calculated result PSNR value.
In step S402, to the reference base picture operating structure in the picture to be detected and the preset reference picture library
Similarity Algorithm obtains structural similarity calculated result.
In the exemplary embodiment, structural similarity (Structural Similarity, referred to as: SSIM) algorithm master
To be used to measure picture structure integrality, be a kind of image quality measure index.In practical application, generally with sliding window to figure
As carrying out piecemeal, sliding window here is generally Gauss window, and with the mean value of each window of Gauss weighted calculation, variance and
Covariance.SSIM algorithm is also a kind of image quality evaluation index referred to entirely, it is respectively from brightness, contrast, structure tripartite
Measure image similarity in face.Its calculation formula is SSIM (X, Y)=L (X, Y) * C (X, Y) * S (X, Y);Wherein, brightness ratio is to letter
Counting formula isContrast contrast function formula isStructure Comparison function
Formula isWherein, uX、uYThe mean value of image X and Y is respectively indicated, The variance of image X and Y is respectively indicated,σX、σYRespectively indicate image
The standard deviation of X and Y,σXYRepresentative image X and Y covariance,C1, C2And C3For constant, to be in order to avoid denominator be 0 and
It maintains to stablize.Usually take C1=(K1* L) ^2, C2=(K2* L) ^2, C3=C2/ 2, generally K1=0.01, K2=0.03, L=255
(L is the dynamic range of pixel value, is generally all taken as 255).The value range of SSIM is [0,1], and the calculated value of SSIM is bigger, table
Diagram image distortion is smaller, i.e., picture to be detected and the similarity degree of the reference base picture in preset reference map valut are higher.Pass through
To the reference base picture operating structure Similarity Algorithm in picture to be detected operation and preset reference picture library, structural similarity is obtained
Calculated result SSIM value.
In step S403, by the Y-PSNR calculated result and the structural similarity calculated result respectively multiplied by
Respective default weight, obtains third similarity.
In the exemplary embodiment, by PSNR calculated value obtained above multiplied by its preset weighted value, then will
The SSIM value arrived is multiplied by its preset weighted value, the product addition that the two obtains, and can access third phase like the numerical value of degree.
By the combination of PSNR algorithm and SSIM algorithm, the situation more unilateral using calculated result caused by single algorithm is avoided,
Improve the precision and accuracy of calculated result.
With continued reference to Fig. 3, after obtaining third similarity, in step s 304, determined according to the third similarity
To the similarity testing result of the picture to be detected.
It should be noted that the comparison accuracy of third alignment algorithm is higher than the precision of second alignment algorithm.
In the exemplary embodiment, the numerical value based on obtained third similarity, it is available to picture to be detected
Similarity testing result can according to testing result be handled picture to be detected in turn.Processing mode may is that for example:
Similar threshold value is e, and dissimilar threshold value is f, then detects picture to be detected and preset reference when the first similarity is less than or equal to e
Reference base picture in picture library is similar, at this point it is possible to which picture to be measured is moved to similar pictures file, completes to figure to be detected
The similarity of piece detects;When the first similarity is more than or equal to f, then detect in picture to be detected and preset reference picture library
Reference base picture is dissimilar, at this point it is possible to call difference algorithm to picture to be detected, difference nomography mainly marks mapping to be checked
Difference between piece and reference base picture, the place bigger to diversity factor are infused with red collimation mark, have the place of different highlighted aobvious
Show do not have discrepant place to carry out translucent processing.It will finally be moved to wrong catalogue to mapping, and generate disparity map simultaneously,
It is put into difference drawings list, completes to detect the similarity of picture to be detected.
The device of the invention embodiment introduced below can be used for executing the above-mentioned picture similarity detection side of the present invention
Method.
Fig. 5 diagrammatically illustrates picture similarity detection apparatus block diagram according to an embodiment of the invention, the detection
Device can be set in the server of picture similarity detection.
Referring to Figure 5, picture similarity detection apparatus block diagram 500 according to an embodiment of the invention, comprising such as
Lower module: the first comparison module 501, the second comparison module 502 are illustrated in detailed below:
First comparison module 501, for obtaining picture to be detected, using the first alignment algorithm by the picture to be detected with
Reference base picture in preset reference map valut is compared, and obtains the first similarity, presets when first similarity meets
Similarity threshold condition when, complete to detect the similarity of the picture to be detected.
In the exemplary embodiment, the first comparison module is used to get the picture to be detected after pretreatment,
And picture to be detected and the reference base picture in preset reference map valut are compared with the perceptual hash algorithm of the first precision
It is right, the first similarity value is obtained, also, when the first similarity value meets preset similarity threshold condition, according to default threshold
Value condition handles similar pictures and dissimilar picture, in addition, for the mapping to be checked for being unsatisfactory for similarity threshold condition
Piece is compared again.
Second comparison module 502, for when first similarity is unsatisfactory for the preset similarity threshold condition,
The picture to be detected and the reference base picture in the preset reference map valut are compared again using the second alignment algorithm, obtained
The second similarity is obtained, and is determined whether again according to second similarity by the picture to be detected and the preset benchmark
Reference base picture in picture library is compared, wherein the comparison accuracy of the second alignment algorithm is higher than first alignment algorithm
Precision.
In the exemplary embodiment, the second comparison module be used for by the first similarity be unsatisfactory for preset threshold condition to
Detection picture is compared with the reference base picture in preset reference map valut with the perceptual hash algorithm of the second precision again,
The second similarity is obtained, and picture to be detected is handled according to the second similarity value.For being unsatisfactory for similarity threshold item
The picture to be detected of part runs third alignment algorithm, the i.e. combination of Y-PSNR algorithm and structural similarity algorithm.Pass through three
The similarity alignment algorithm of a level improves the precision and accuracy rate of similarity detection.
Due to each functional module and above-mentioned picture phase of the picture similarity detection apparatus of example embodiments of the present invention
It is corresponding like the step of the example embodiment for spending detection method, therefore for undisclosed details in apparatus of the present invention embodiment, it asks
Referring to the embodiment of the above-mentioned picture similarity detection method of the present invention.
Below with reference to Fig. 6, it illustrates the computer systems 600 for the electronic equipment for being suitable for being used to realize the embodiment of the present invention
Structural schematic diagram.The computer system 600 of electronic equipment shown in Fig. 6 is only an example, should not be to the embodiment of the present invention
Function and use scope bring any restrictions.
As shown in fig. 6, computer system 600 includes central processing unit (CPU) 601, it can be read-only according to being stored in
Program in memory (ROM) 602 or be loaded into the program in random access storage device (RAM) 603 from storage section 608 and
Execute various movements appropriate and processing.In RAM 603, it is also stored with various programs and data needed for system operatio.CPU
601, ROM 602 and RAM 603 is connected with each other by bus 604.Input/output (I/O) interface 605 is also connected to bus
604。
I/O interface 605 is connected to lower component: the importation 606 including keyboard, mouse etc.;It is penetrated including such as cathode
The output par, c 607 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.;Storage section 608 including hard disk etc.;
And the communications portion 609 of the network interface card including LAN card, modem etc..Communications portion 609 via such as because
The network of spy's net executes communication process.Driver 610 is also connected to I/O interface 605 as needed.Detachable media 611, such as
Disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on as needed on driver 610, in order to read from thereon
Computer program be mounted into storage section 608 as needed.
Particularly, according to an embodiment of the invention, may be implemented as computer above with reference to the process of flow chart description
Software program.For example, the embodiment of the present invention includes a kind of computer program product comprising be carried on computer-readable medium
On computer program, which includes the program code for method shown in execution flow chart.In such reality
It applies in example, which can be downloaded and installed from network by communications portion 609, and/or from detachable media
611 are mounted.When the computer program is executed by central processing unit (CPU) 601, executes and limited in the system of the application
Above-mentioned function.
It should be noted that computer-readable medium shown in the present invention can be computer-readable signal media or meter
Calculation machine readable storage medium storing program for executing either the two any combination.Computer readable storage medium can be for example but not limited to
Electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination.It is computer-readable
The more specific example of storage medium can include but is not limited to: have electrical connection, the portable computing of one or more conducting wires
Machine disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable programmable read only memory (EPROM
Or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device or above-mentioned
Any appropriate combination.In the present invention, computer readable storage medium can be it is any include or storage program it is tangible
Medium, the program can be commanded execution system, device or device use or in connection.And in the present invention,
Computer-readable signal media may include in a base band or as the data-signal that carrier wave a part is propagated, wherein carrying
Computer-readable program code.The data-signal of this propagation can take various forms, and including but not limited to electromagnetism is believed
Number, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be computer-readable storage medium
Any computer-readable medium other than matter, the computer-readable medium can be sent, propagated or transmitted for being held by instruction
Row system, device or device use or program in connection.The program code for including on computer-readable medium
It can transmit with any suitable medium, including but not limited to: wireless, electric wire, optical cable, RF etc. or above-mentioned any conjunction
Suitable combination.
Flow chart and block diagram in attached drawing are illustrated according to the system of various embodiments of the invention, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of above-mentioned module, program segment or code include one or more
Executable instruction for implementing the specified logical function.It should also be noted that in some implementations as replacements, institute in box
The function of mark can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are practical
On can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it wants
It is noted that the combination of each box in block diagram or flow chart and the box in block diagram or flow chart, can use and execute rule
The dedicated hardware based systems of fixed functions or operations is realized, or can use the group of specialized hardware and computer instruction
It closes to realize.
Being described in unit involved in the embodiment of the present invention can be realized by way of software, can also be by hard
The mode of part realizes that described unit also can be set in the processor.Wherein, the title of these units is in certain situation
Under do not constitute restriction to the unit itself.
As on the other hand, present invention also provides a kind of computer-readable medium, which be can be
Included in electronic equipment described in above-described embodiment;It is also possible to individualism, and without in the supplying electronic equipment.
Above-mentioned computer-readable medium carries one or more program, when the electronics is set by one for said one or multiple programs
When standby execution, so that the electronic equipment realizes such as above-mentioned picture similarity detection method as described in the examples.
For example, the electronic equipment may be implemented as shown in Figure 1: step S101 obtains picture to be detected, utilizes
The picture to be detected is compared first alignment algorithm with the reference base picture in preset reference map valut, obtains the first phase
Like degree;Step S102 is completed when first similarity meets preset similarity threshold condition to the picture to be detected
Similarity detection;Step S103 utilizes when first similarity is unsatisfactory for the preset similarity threshold condition
Two alignment algorithms compare the picture to be detected and the reference base picture in the preset reference map valut again, obtain second
Similarity, and determined whether again according to second similarity by the picture to be detected and the preset reference map valut
In reference base picture be compared;Wherein, the comparison accuracy of the second alignment algorithm is higher than the precision of first alignment algorithm.
For another example, each step as shown in any figure of Fig. 2-5 may be implemented in the electronic equipment.
It should be noted that although being referred to several modules or list for acting the equipment executed in the above detailed description
Member, but this division is not enforceable.In fact, embodiment according to the present invention, it is above-described two or more
Module or the feature and function of unit can embody in a module or unit.Conversely, an above-described mould
The feature and function of block or unit can be to be embodied by multiple modules or unit with further division.
Through the above description of the embodiments, those skilled in the art is it can be readily appreciated that example described herein is implemented
Mode can also be realized by software realization in such a way that software is in conjunction with necessary hardware.Therefore, according to the present invention
The technical solution of embodiment can be embodied in the form of software products, which can store non-volatile at one
Property storage medium (can be CD-ROM, USB flash disk, mobile hard disk etc.) in or network on, including some instructions are so that a calculating
Equipment (can be personal computer, server, touch control terminal or network equipment etc.) executes embodiment according to the present invention
Method.
Those skilled in the art after considering the specification and implementing the invention disclosed here, will readily occur to of the invention its
Its embodiment.This application is intended to cover any variations, uses, or adaptations of the invention, these modifications, purposes or
Person's adaptive change follows general principle of the invention and including the undocumented common knowledge in the art of the present invention
Or conventional techniques.The description and examples are only to be considered as illustrative, and true scope and spirit of the invention are by following
Claim is pointed out.
It should be understood that the present invention is not limited to the precise structure already described above and shown in the accompanying drawings, and
And various modifications and changes may be made without departing from the scope thereof.The scope of the present invention is limited only by the attached claims.
Claims (10)
1. a kind of picture similarity detection method characterized by comprising
Picture to be detected is obtained, using the first alignment algorithm by the benchmark in the picture to be detected and preset reference map valut
Picture is compared, and obtains the first similarity;
When first similarity meets preset similarity threshold condition, complete to examine the similarity of the picture to be detected
It surveys;
When first similarity is unsatisfactory for the preset similarity threshold condition, using the second alignment algorithm will it is described to
Detection picture compares again with the reference base picture in the preset reference map valut, obtains the second similarity, and according to described
Second similarity determines whether again to carry out the reference base picture in the picture to be detected and the preset reference map valut
It compares;
Wherein, the comparison accuracy of the second alignment algorithm is higher than the precision of first alignment algorithm.
2. picture similarity detection method according to claim 1, which is characterized in that determined according to second similarity
Whether the picture to be detected is compared with the reference base picture in the preset reference map valut again, comprising:
Judge whether second similarity meets the preset similarity threshold condition;
When second similarity meets the preset similarity threshold condition, complete to the similar of the picture to be detected
Degree detection;
When second similarity is unsatisfactory for preset condition, using third alignment algorithm by the picture to be detected with it is described pre-
If reference map valut in reference base picture compare again, obtain third similarity;
The similarity testing result to the picture to be detected is determined according to the third similarity;
Wherein, the comparison accuracy of third alignment algorithm is higher than the precision of second alignment algorithm.
3. picture similarity detection method according to claim 2, which is characterized in that
First alignment algorithm is the picture hash algorithm of the first precision, and second alignment algorithm is the picture of the second precision
Hash algorithm, the third alignment algorithm are the combination of Y-PSNR algorithm and structural similarity algorithm.
4. picture similarity detection method according to any one of claims 1 to 3, which is characterized in that compared using first
The picture to be detected is compared algorithm with the reference base picture in preset reference map valut, obtains the first similarity, packet
It includes:
Reference base picture in the picture to be detected and the preset reference map valut is narrowed down into same order, and is converted
For corresponding gray scale picture;
Discrete cosine transform is carried out to the gray scale picture and obtains coefficient matrix;
The coefficient matrix is handled, picture fingerprint to be detected and reference base picture fingerprint are generated;
The picture fingerprint to be detected and the reference base picture fingerprint are compared, the first similarity is obtained.
5. picture similarity detection method according to any one of claims 1 to 3, which is characterized in that compared using second
Algorithm compares the picture to be detected and the reference base picture in the preset reference map valut again, and it is similar to obtain second
Degree, comprising:
The picture to be detected is split with the reference base picture in the preset reference map valut, generates corresponding at least one
Open segmentation picture;
First alignment algorithm is run at least one segmentation picture again, obtains the second similarity.
6. picture similarity detection method according to claim 2 or 3, which is characterized in that utilize third alignment algorithm will
The picture to be detected compares again with the reference base picture in the preset reference map valut, obtains third similarity, comprising:
Y-PSNR algorithm is run to the reference base picture in the picture to be detected and the preset reference picture library, obtains peak
It is worth signal-to-noise ratio computation result;
To the reference base picture operating structure Similarity Algorithm in the picture to be detected and the preset reference picture library, tied
Structure Similarity measures result;
By the Y-PSNR calculated result and the structural similarity calculated result respectively multiplied by respective default weight, obtain
To third similarity.
7. picture similarity detection method according to any one of claims 1 to 3, obtains picture to be detected, comprising:
Edge detection is carried out to original image based on edge detection algorithm, obtains the outer layer coordinate of original image;
It is cut based on the outer layer coordinate pair original image, obtains picture to be detected.
8. a kind of picture similarity detection apparatus characterized by comprising
First comparison module, for obtaining picture to be detected, using the first alignment algorithm by the picture to be detected with it is preset
Reference base picture in reference map valut is compared, obtain the first similarity, when first similarity meet it is preset similar
When spending threshold condition, complete to detect the similarity of the picture to be detected;
Second comparison module, for utilizing when first similarity is unsatisfactory for the preset similarity threshold condition
Two alignment algorithms compare the picture to be detected and the reference base picture in the preset reference map valut again, obtain second
Similarity, and determined whether again according to second similarity by the picture to be detected and the preset reference map valut
In reference base picture be compared;
Wherein, the comparison accuracy of the second alignment algorithm is higher than the precision of first alignment algorithm.
9. a kind of computer-readable medium, is stored thereon with computer program, which is characterized in that described program is executed by processor
The method of picture similarity detection of the Shi Shixian as described in any one of claims 1 to 7.
10. a kind of electronic equipment characterized by comprising
One or more processors;
Storage device, for storing one or more programs, when one or more of programs are by one or more of processing
When device executes, so that one or more of processors realize the picture similarity inspection as described in any one of claims 1 to 7
The method of survey.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811270419.3A CN109447154B (en) | 2018-10-29 | 2018-10-29 | Picture similarity detection method, device, medium and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811270419.3A CN109447154B (en) | 2018-10-29 | 2018-10-29 | Picture similarity detection method, device, medium and electronic equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109447154A true CN109447154A (en) | 2019-03-08 |
CN109447154B CN109447154B (en) | 2021-06-04 |
Family
ID=65548614
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811270419.3A Active CN109447154B (en) | 2018-10-29 | 2018-10-29 | Picture similarity detection method, device, medium and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109447154B (en) |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110399511A (en) * | 2019-07-23 | 2019-11-01 | 中南民族大学 | Image cache method, equipment, storage medium and device based on Redis |
CN110503504A (en) * | 2019-03-14 | 2019-11-26 | 杭州海康威视数字技术股份有限公司 | Information identifying method, device and the equipment of networking products |
CN110610485A (en) * | 2019-08-26 | 2019-12-24 | 北京许继电气有限公司 | Ultra-high voltage transmission line channel hidden danger early warning method based on SSIM algorithm |
CN110688514A (en) * | 2019-08-30 | 2020-01-14 | 中国人民财产保险股份有限公司 | Insurance claim settlement image data duplicate checking method and device |
CN111143601A (en) * | 2019-12-31 | 2020-05-12 | 深圳市芭田生态工程股份有限公司 | Image processing method |
CN111178437A (en) * | 2019-12-30 | 2020-05-19 | 深圳市芭田生态工程股份有限公司 | Fertilizing method and computer readable storage medium |
CN111242053A (en) * | 2020-01-16 | 2020-06-05 | 国网山西省电力公司电力科学研究院 | Power transmission line flame detection method and system |
CN111275096A (en) * | 2020-01-17 | 2020-06-12 | 青梧桐有限责任公司 | Homonymous cell identification method and system based on image identification |
CN111538503A (en) * | 2020-03-27 | 2020-08-14 | 北京达佳互联信息技术有限公司 | Installation package volume optimization method and device, electronic equipment and storage medium |
CN111913873A (en) * | 2020-06-17 | 2020-11-10 | 浙江数链科技有限公司 | Picture verification method, device and system and computer readable storage medium |
CN111966600A (en) * | 2020-08-31 | 2020-11-20 | 平安健康保险股份有限公司 | Webpage testing method and device, computer equipment and computer readable storage medium |
CN112183342A (en) * | 2020-09-28 | 2021-01-05 | 国网安徽省电力有限公司检修分公司 | Comprehensive convertor station defect identification method with template |
CN112365487A (en) * | 2020-11-20 | 2021-02-12 | 中国邮政集团有限公司广州市分公司 | Stamp identification method, system, device and storage medium |
CN112561986A (en) * | 2020-12-02 | 2021-03-26 | 南方电网电力科技股份有限公司 | Secondary alignment method, device, equipment and storage medium for inspection robot holder |
CN113139589A (en) * | 2021-04-12 | 2021-07-20 | 网易(杭州)网络有限公司 | Picture similarity detection method and device, processor and electronic device |
CN113435485A (en) * | 2021-06-15 | 2021-09-24 | 北京百度网讯科技有限公司 | Picture detection method and device, electronic equipment and storage medium |
CN116225972A (en) * | 2023-05-09 | 2023-06-06 | 成都赛力斯科技有限公司 | Picture difference comparison method, device and storage medium |
Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102013018A (en) * | 2010-12-03 | 2011-04-13 | 黄晓峰 | Closed loop image comparison method |
CN103336776A (en) * | 2013-05-13 | 2013-10-02 | 云南瑞攀科技有限公司 | Image searching method based on image content |
CN103903249A (en) * | 2012-12-27 | 2014-07-02 | 纽海信息技术(上海)有限公司 | Image matching system and method |
CN104112116A (en) * | 2011-06-30 | 2014-10-22 | 深圳市君盛惠创科技有限公司 | Cloud server |
CN104318259A (en) * | 2014-10-20 | 2015-01-28 | 北京齐尔布莱特科技有限公司 | Target picture identifying device and method for and computing device |
CN105117724A (en) * | 2015-07-30 | 2015-12-02 | 北京邮电大学 | License plate positioning method and apparatus |
CN105631487A (en) * | 2015-12-31 | 2016-06-01 | 北京奇艺世纪科技有限公司 | Image comparison method, device, video comparison method and device |
CN106372606A (en) * | 2016-08-31 | 2017-02-01 | 北京旷视科技有限公司 | Target object information generation method and unit identification method and unit and system |
CN106650829A (en) * | 2017-01-04 | 2017-05-10 | 华南理工大学 | Picture similarity calculation method |
US20170140212A1 (en) * | 2015-11-16 | 2017-05-18 | MorphoTrak, LLC | Facial Matching System |
CN106845397A (en) * | 2017-01-18 | 2017-06-13 | 湘潭大学 | A kind of confirming face method based on measuring similarity |
CN107844803A (en) * | 2017-10-30 | 2018-03-27 | 中国银联股份有限公司 | The method and apparatus that a kind of picture compares |
-
2018
- 2018-10-29 CN CN201811270419.3A patent/CN109447154B/en active Active
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102013018A (en) * | 2010-12-03 | 2011-04-13 | 黄晓峰 | Closed loop image comparison method |
CN104112116A (en) * | 2011-06-30 | 2014-10-22 | 深圳市君盛惠创科技有限公司 | Cloud server |
CN103903249A (en) * | 2012-12-27 | 2014-07-02 | 纽海信息技术(上海)有限公司 | Image matching system and method |
CN103336776A (en) * | 2013-05-13 | 2013-10-02 | 云南瑞攀科技有限公司 | Image searching method based on image content |
CN104318259A (en) * | 2014-10-20 | 2015-01-28 | 北京齐尔布莱特科技有限公司 | Target picture identifying device and method for and computing device |
CN105117724A (en) * | 2015-07-30 | 2015-12-02 | 北京邮电大学 | License plate positioning method and apparatus |
US20170140212A1 (en) * | 2015-11-16 | 2017-05-18 | MorphoTrak, LLC | Facial Matching System |
CN105631487A (en) * | 2015-12-31 | 2016-06-01 | 北京奇艺世纪科技有限公司 | Image comparison method, device, video comparison method and device |
CN106372606A (en) * | 2016-08-31 | 2017-02-01 | 北京旷视科技有限公司 | Target object information generation method and unit identification method and unit and system |
CN106650829A (en) * | 2017-01-04 | 2017-05-10 | 华南理工大学 | Picture similarity calculation method |
CN106845397A (en) * | 2017-01-18 | 2017-06-13 | 湘潭大学 | A kind of confirming face method based on measuring similarity |
CN107844803A (en) * | 2017-10-30 | 2018-03-27 | 中国银联股份有限公司 | The method and apparatus that a kind of picture compares |
Non-Patent Citations (2)
Title |
---|
A HARIS RANGKUTI 等: "Analysis of image similarity with CBIR concept using wavelet transform and threshold algorithm", 《2013 IEEE SYMPOSIUM ON COMPUTERS & INFORMATICS (ISCI)》 * |
郑德忠,崔法毅: "小波包特征提取及方差相似度的人脸识别", 《光学技术》 * |
Cited By (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110503504B (en) * | 2019-03-14 | 2022-02-15 | 杭州海康威视数字技术股份有限公司 | Information identification method, device and equipment of network product |
CN110503504A (en) * | 2019-03-14 | 2019-11-26 | 杭州海康威视数字技术股份有限公司 | Information identifying method, device and the equipment of networking products |
CN110399511A (en) * | 2019-07-23 | 2019-11-01 | 中南民族大学 | Image cache method, equipment, storage medium and device based on Redis |
CN110610485A (en) * | 2019-08-26 | 2019-12-24 | 北京许继电气有限公司 | Ultra-high voltage transmission line channel hidden danger early warning method based on SSIM algorithm |
CN110688514A (en) * | 2019-08-30 | 2020-01-14 | 中国人民财产保险股份有限公司 | Insurance claim settlement image data duplicate checking method and device |
CN111178437A (en) * | 2019-12-30 | 2020-05-19 | 深圳市芭田生态工程股份有限公司 | Fertilizing method and computer readable storage medium |
CN111143601A (en) * | 2019-12-31 | 2020-05-12 | 深圳市芭田生态工程股份有限公司 | Image processing method |
CN111242053A (en) * | 2020-01-16 | 2020-06-05 | 国网山西省电力公司电力科学研究院 | Power transmission line flame detection method and system |
CN111275096A (en) * | 2020-01-17 | 2020-06-12 | 青梧桐有限责任公司 | Homonymous cell identification method and system based on image identification |
CN111538503A (en) * | 2020-03-27 | 2020-08-14 | 北京达佳互联信息技术有限公司 | Installation package volume optimization method and device, electronic equipment and storage medium |
CN111913873A (en) * | 2020-06-17 | 2020-11-10 | 浙江数链科技有限公司 | Picture verification method, device and system and computer readable storage medium |
CN111966600A (en) * | 2020-08-31 | 2020-11-20 | 平安健康保险股份有限公司 | Webpage testing method and device, computer equipment and computer readable storage medium |
CN111966600B (en) * | 2020-08-31 | 2023-08-04 | 平安健康保险股份有限公司 | Webpage testing method, webpage testing device, computer equipment and computer readable storage medium |
CN112183342A (en) * | 2020-09-28 | 2021-01-05 | 国网安徽省电力有限公司检修分公司 | Comprehensive convertor station defect identification method with template |
CN112183342B (en) * | 2020-09-28 | 2022-07-12 | 国网安徽省电力有限公司检修分公司 | Comprehensive convertor station defect identification method with template |
CN112365487A (en) * | 2020-11-20 | 2021-02-12 | 中国邮政集团有限公司广州市分公司 | Stamp identification method, system, device and storage medium |
CN112561986A (en) * | 2020-12-02 | 2021-03-26 | 南方电网电力科技股份有限公司 | Secondary alignment method, device, equipment and storage medium for inspection robot holder |
CN113139589A (en) * | 2021-04-12 | 2021-07-20 | 网易(杭州)网络有限公司 | Picture similarity detection method and device, processor and electronic device |
CN113139589B (en) * | 2021-04-12 | 2023-02-28 | 网易(杭州)网络有限公司 | Picture similarity detection method and device, processor and electronic device |
CN113435485A (en) * | 2021-06-15 | 2021-09-24 | 北京百度网讯科技有限公司 | Picture detection method and device, electronic equipment and storage medium |
CN116225972A (en) * | 2023-05-09 | 2023-06-06 | 成都赛力斯科技有限公司 | Picture difference comparison method, device and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN109447154B (en) | 2021-06-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109447154A (en) | Picture similarity detection method, device, medium and electronic equipment | |
CN108229296B (en) | Face skin attribute identification method and device, electronic equipment and storage medium | |
US9069948B2 (en) | Methods, systems, and media for measuring quality of gesture-based passwords | |
CN108898086A (en) | Method of video image processing and device, computer-readable medium and electronic equipment | |
CN109564575A (en) | Classified using machine learning model to image | |
CN108846440A (en) | Image processing method and device, computer-readable medium and electronic equipment | |
CN109558779A (en) | Image detecting method and device | |
CN111915480A (en) | Method, apparatus, device and computer readable medium for generating feature extraction network | |
CN114494784A (en) | Deep learning model training method, image processing method and object recognition method | |
CN110390706A (en) | A kind of method and apparatus of object detection | |
CN114092963A (en) | Key point detection and model training method, device, equipment and storage medium | |
CN113705362A (en) | Training method and device of image detection model, electronic equipment and storage medium | |
CN109711441A (en) | Image classification method, device, storage medium and electronic equipment | |
CN115311469A (en) | Image labeling method, training method, image processing method and electronic equipment | |
CN112241761A (en) | Model training method and device and electronic equipment | |
CN110717405A (en) | Face feature point positioning method, device, medium and electronic equipment | |
CN110008974A (en) | Behavioral data prediction technique, device, electronic equipment and computer storage medium | |
CN115457365A (en) | Model interpretation method and device, electronic equipment and storage medium | |
CN114707638A (en) | Model training method, model training device, object recognition method, object recognition device, object recognition medium and product | |
CN109948636A (en) | Data fusion method and device | |
CN117693754A (en) | Training masked automatic encoders for image restoration | |
CN112861940A (en) | Binocular disparity estimation method, model training method and related equipment | |
CN113298747A (en) | Picture and video detection method and device | |
CN107493471B (en) | A kind of calculation method and device of video transmission quality | |
CN112348615A (en) | Method and device for auditing information |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |